Potash from lithium brines

SQM, as Eoote, initially selected a brine extraction location for its well field where the brine had the maximum potassium and the least sulfate for potash and lithium production, and later a location with the maximum sulfate content for potassium sulfate production (Fig. 1.57). Because of this the plants could initially use the simplest processes and have the lowest capital and operating costs. In the initial operation brine with up to 3400 ppm Li was pumped from the Salar in 40 wells, 28 m deep on a 200-500 m grid, which delivered up to 5280 m /hr of brine to the solar ponds. There were also 13 monitoring wells to follow any changes in the brine concentration and its depth from the surface. The ponds were lined with flexible PVC or reinforced hypalon membranes, and the brine flowed through the sections of the pond system in series. The initial salt ponds had an area of 1.16 million m followed by 3.36 million m for the sylvinite ponds, and later 1 million m of ponds were installed for lithium production. The plant employed 184 people, of which 120 were hired from the sparsely populated local area. Contractors were used to drill and maintain the weUs, harvest the salts, transport them to their respective stockpiles, and reclaim the sylvinite to feed the potash plant s conveyor belt. They also provided all of the miscellaneous trucking needed at the Salar, and transported the potash to Coya Sur or Maria Elena and the concentrated lithium chloride brine to the Salar de Carmen. SQM unloaded the brine and potash, and stacked the later material at its nitrate plants (Harben and Edwards, 1997). 

Precipitating lithium from low-lithium brines with sodium phosphate has also been tested, after the model of licons being precipitated from Searles Lake brine. Tandy and Canfy (1993) smdied the precipitation of lithium phosphate from Dead Sea potash pond end-liquor, and found that perhaps a 70% Li recovery could be obtained. By adding over a 30-fold molar excess of disodium phosphate to the lithium in the brine, adjusting the pH to 6-7, heating to 80°C, and with a 20-30 min residence time about 76% of the lithium would be precipitated along with dicalcium phosphate and the excess disodium phosphate. The precipitate contained about 0.3% Li, and could be leached with water to recover over 90% of the Li, with the remainder being in the residual phosphate precipitate. The filtrate contained about 1440 ppm Li in a sodium phosphate-chloride solution (Table 1.34). 

The Dead Sea is one of the world s largest and lowest inland lakes, containing a concentrated calcium-magnesium-sodium-potassium chloride brine, with about 10 ppm Li (Table 1.9) and reserves of about 2 milUon tons of Li. The brine is commercially evaporated in large solar ponds to produce potash in both Israel and Jordan, and their pond end-liquors often contain about 30 ppm Li. Some of this brine is processed for bromine and magnesia recovery, but most of it is merely returned to the sea. Because of its ready availability and potential value several laboratory studies have been made on lithium recovery from it, but without economic success. 

In 1960 the industry s over-capacity only allowed operation at about 20% of capacity, and Maywood ceased the production of lithium compounds. LCA closed their Minneapolis plant in 1959 (and canceled their long-term ore contract with Quebec Lithium), while American Potash Chemical Co. closed their Texas plant in 1963. American Potash s Searles Lake lithium operation had started in 1951 and closed in 1978. Quebec Lithium in turn started producing lithium chemicals, but closed their plant in 1965. The production of ore from South Dakota stopped in 1969, and sanctions against Southern Rhodesia (still one of the world s major suppliers) curtailed their ore imports from 1965-1980. Foote s Clayton Valley brine operation commenced in 1966, and LCA started mining spodumene in North Carolina in 1968. Foote s Salar de Atacama operation started in 1984, while SQM s started at the Salar de Atacama, and FMC s (formerly LCA) at the Salar de Hombre Muerto in 1997. Both of the North Carolina mines closed after their brine operations had been well established, and FMC essentially closed their Hombre Muerto plant in 1998 because of SQM s greatly reduced lithium carbonate pricing. 

Brine from the sylvinite ponds next went to the camallite ponds (Fig. 1.64), and from there to the 500,000 m bischoffite ponds (Fig. 1.65). These two series of lithium ponds were also periodically taken out of service to harvest predominantly camallite from the first ponds, and bischoffite from the later ponds. These minerals were stockpiled separately, with some of the bischoffite sold as magnesium chloride (with a capacity of 450,000 mt/yr), and the camallite saved for later conversion to potash. The six camallite ponds were divided into two groups of three, with the higher sulfate brine directed to one group, and then its end-liquor was returned to the Salar by being flooded onto its porous surface. The final brine from the bischoffite ponds contained 6.0-6.1% Li, and was sent to 40,000 m, about 3 m deep holding ponds to await tmck shipment to the lithium carbonate plant. The plant had a capacity of 22,000 mt/yr of Li2C03 in 2002, to be raised to 28,000 mt/yr in 2003 (Moura, 2002 Etchart, 2002 Nakousi, 2003). 

The Michigan Basin brines very low pH helps to explain their ability to leach and react with other rocks, as is indicated by their high contents of strontium, barium and other metals, although much of the Sr and Ba probably came from the reaction with calcite. Geothermal water also probably mixed with some of the formations, as indicated by the variable presence of iodine, boron, lithium, cesium, rubidium and other rare metals. With most of the brines, the calcium concentration is somewhat higher than its magnesium equivalent in seawater end liquor from a potash deposit, and the potassium a little lower. Wilson and Long (1993) speculated that this occurred by the conversion of the clays kaolinite and smectite to illite ... 

---